Eli Carter wrote:
> I'm dealing with an AMD chip that does not have the station address in
> the PROM at the base address, but resides in the "Physical Address
> Registers" in the chip (thanks to the bootloader in my case).  This
> patch makes the driver try those registers if the station address read
> from the PROM is 00:00:00:00:00:00.
> I think others dealing with similar setups may find this helpful.  The
> other lance-derived drivers might benefit from a similar patch, but I
> don't have that hardware for testing.

<aside>
Added Peter since he's given feedback on a past pcnet32 patch.
</aside>

Two patches, one for 2.2.18 (patch-pcnet32-mac22), and one for 2.4.1
(patch-pcnet32-mac24) which should each apply cleanly.

Changes:
- Moved address validity check to a function.  (Should this go somewhere
other drivers can call it?)
- Check the second address and set the address to 00:00:00:00:00:00 if
it fails
- Check the address at open time as well, failing with -EINVAL.

I think that should address Alan's initial feedback.

What else?

TIA,

Eli
--------------------.              Rule of Accuracy: When working toward
Eli Carter          |               the solution of a problem, it always 
eli.carter@inet.com `--------------------- helps if you know the answer.